YvanYin/VNL_Monocular_Depth_Prediction

Depth in meters

utsavpatel22 opened this issue · 4 comments

Hello, Thank you so much for sharing this code. I am working on a project where we need to get the pedestrian velocities. To estimate the velocities, I need the depth information. We are using a depth camera for real-world implementation, but I need to run the same code on a dataset of monocular images to evaluate our algorithm. I am using your code to get the depth frames, but I am not sure about the unit of depth in the output depth frames. Is there a way to get the depth in meters?

Hi, which model do you use?

Hi @YvanYin, I am using ResNext101_32x4d backbone, trained on NYU dataset model. Will there be a difference in the estimated depth for different models?

Yes, there are some differences. When train the model on NYU, the farthest depth is 10m, while it is 80m on KITTI.

Hi, @YvanYin ,
Is it true that the depth range in kitti is 80m? I find out that the pred_depth_scale multiply 60000 in the test_any_images.py. Is it possible that the depth range in kitti is 60m?
图片